PRODUCTION OF ANDROGENETIC BARLEY DOUBLED HAPLOID LINES1

ANA LÚCIA STIVAL DA SILVA2, MARIA IRENE BAGGIO DE MORAES FERNANDES3, GERARDO ARIAS4,ALFREDO GUI FERREIRA5 and JOÃO CARLOS HAAS6

ABSTRACT - Anther culture allows obtention of homozygous lines in one generation instead of seven required by conventional breeding programs. In barley (Hordeum vulgare L.) this technique is well-established, but response is greatly influenced by genotype and growing conditions of donor plants. Therefore, an important goal is to adapt media and methods, aiming to satisfy the specific requirements of every material one has been working with. In this paper androgenetic capacity of F1 hybrids is evaluated and an efficient protocol for regeneration of barley doubled haploids is established in order to make possible their utilization in breeding programs. Two culture media, modified N6 and MS, were employed for induction of androgenesis. Frequency of responsive anthers, green plant regeneration, albinism, and spontaneous doubling were evaluated. Results show that average percentage of responsive anthers was greater in N6 (30.32%) than in MS (6.39%) medium. There was a considerable influence of the genotype for all traits. A total of 192 doubled haploid lines were obtained from different crosses. These lines set seed and were multiplied for agronomic testing in the field.

Index terms: anther culture.

PRODUÇÃO DE LINHAGENS DUPLO-HAPLÓIDES ANDROGENÉTICAS DE CEVADA

RESUMO - A cultura de anteras permite a obtenção de linhagens homozigotas em uma geração, em vez das sete requeridas pelos programas convencionais de melhoramento. Essa técnica encontra-se bem estabelecida com respeito a cevada (Hordeum vulgare L.), embora a resposta seja muito influenciada pelo genótipo e pelas condições de cultivo das plantas doadoras. Por isso, torna-se importante otimizar meios e métodos, buscando satisfazer os requerimentos específicos de cada material usado no trabalho. O objetivo deste trabalho foi avaliar a capacidade androgenética de híbridos F 1 de cevada e estabelecer um protocolo eficiente de regeneração de duplo-haplóides, para tornar possível sua utilização em programas de melhoramento. Dois meios de cultura, N6 e MS modificados, foram utilizados para indução de androgênese. Avaliou-se a freqüência de anteras responsivas e de regeneração de plantas verdes, e a porcentagem de albinismo e de duplicação espontânea. Os resultados mostram que a porcentagem média de anteras responsivas no meio N6 (30,32%) foi maior do que no meio MS (6,39%). Houve uma considerável influência do genótipo em relação a todas as variáveis analisadas. No total, 192 genótipos duplo-haplóides, originados de diferentes cruzamentos, foram cultivados até a produção de grãos. As sementes de cada genótipo foram multiplicadas, para avaliação agronômica.

Termos para indexação: cultura de anteras.

1 Accepted for publication on May 5, 1997.2 Biologist, M.Sc., Bolsista da Embrapa-Centro Nacional de

Pesquisa de Trigo (CNPT), Caixa Postal 569, CEP 99001-970 Passo Fundo, RS, Brazil.

3 Naturalist, Drª, Embrapa-CNPT.4 Agronomist, Dr., Embrapa-CNPT.5 Naturalist, Dr., Prof. Titular, CPG/Botânica, Universidade

Federal do Rio Grande do Sul, Av. Paulo Gama, 40, CEP 90046-900 Porto Alegre, RS, Brazil.

6 Agronomist, M.Sc., Embrapa-CNPT.

INTRODUCTION

Since Guha & Maheshwari (1966) first described haploid embryos in Datura innoxia anthers, many groups in different parts of the world have been involved in the production of doubled haploid plants, which are very useful in breeding programs of cultivated species. Classical breeding combines useful traits, like better productivity or adaptation, by crossing different varieties. Nevertheless, in the case of autogamous species, releasing a new cultivar depends on genetic homozygosis, which requires seven to nine generations, after the original cross. Few more years are spent in field trials, making up 10 to 14 years to get a new variety (Fernandes, 1987). Modifications in either the environment or consumer demands may occur before a new

cultivar is released in the market. Therefore, techniques that reduce the costs necessary to release new cultivars are always welcome. In this context, doubled haploid production by anther culture reveals itself as an useful tool. Unlike conventional breeding programs, in which numerous cycles of planting and harvesting segregating populations are necessary, regeneration of fertile plants from gametophytic cells, without fertilization, allows the rapid development of pure lines. According to Fernandes (1987, 1990) and Fernandes et al. (1990), the time necessary to attain homozygosis is shortened from seven to one generation by the production of doubled haploids, and selection becomes more simple and efficient.

The first protocol for barley anther culture was developed by Clapham (1973). At that time frequency of regeneration was very low and almost all regenerated plantlets were albino. Over the last twenty years the number of green plants has been improved by 100-1000 times (Huang & Sunderland, 1982; Olsen, 1987; Shell Internationale Research Maatschappij B.V., 1987; Finnie et al., 1989; Kühlmann & Foroughi-Wehr, 1989; Cai et al., 1992). Numerous doubled haploid lines have been produced and field tested, including some resistant to barley yellow mosaic virus (Foroughi-Wehr & Friedt, 1984). According to Wenzel (1992), barley is one of the crops in which haploid production is well-established. Nevertheless, most of the studies have been done using a small number of responsive cultivars, especially Igri, which has an extraordinary androgenetic capacity. The use of highly responsive genotypes may adapt the technique to genotypes which are not very useful for practical purposes and restrict the genetic variability available to the breeder (Luckett & Smithard, 1992). In order to prevent that it is recommended to develop protocols and culture media applicable to a broader spectrum of genotypes. This question is particularly important when considering the use of anther culture in Brazilian barley breeding programs. It is important to verify if methods developed in Europe and/or in the United States of America are applicable to local genotypes and environmental conditions. Up to now, little is known about the androgenetic capacity of Brazilian varieties and cultivars. The objective of this work was to evaluate androgenetic capacity of hybrid genotypes, in two culture media, and establish a protocol for obtainment of barley doubled-haploids with relevant agronomic characteristics.

MATERIAL AND METHODS

The work was performed on eleven F1 populations, namely AF 279 / PFC 9104, BR 2 / MN 607, BR 2 / PFC 9104, MN 607 / PFC 9104, DEFRA / MN 607, MN 656 / PFC 9104, MN 668 / PFC 9104, PFC 85107 / PFC 9104,PFC 86104 / PFC 9104, PFC 9104 / PFC 9134 and, PFC 9112 / PFC 88154 derived from crosses made at Embrapa-Centro Nacional de Pesquisa de Trigo (National Research Center for Wheat), in Passo Fundo, Rio Grande do Sul, Brazil (28° 15' S, 52 o 24' W, 687 m). Plants were cultivated in pots, in the greenhouse, with ambient light and temperature, during the winter. Spikes containing pollen grains at mid-uninucleate stage were harvested, sterilized with ethanol 96°, and stored in the dark, at 4oC, as described by Huang & Sunderland (1982). Pollen grain development stage was estimated through the distance from the base of the flag-leaf and the anterior leaf. For tested genotypes and existing environmental conditions, such distance was about 5 cm.

After ten days of cold-pretreatment, anthers were cultured in two different culture media: modified MS (Murashige & Skoog, 1962) or N6 (Chu, 1981). Both media contained 60 g/L maltose (Merck/Mikrobiologie), 8 g/L agarose (Sigma, Low melting point, Type VII), 2 mg/L α-Naphthaleneacetic Acid (NAA), and 1 mg/L 6-Benzylaminopurine (BA). Anthers were incubated in the dark, at 25 ± 1oC, for 30 days and Petri dishes were then transferred to the light (60 µE m-2

s-1, 12 h). The frequency of responsive anthers, green and albino plantlets was assessed. Green plantlets were rooted in test tubes containing modified MS medium (Olsen, 1987), with 30 g/L sucrose, 7 g/L agar (Difco, Bacto-agar), and 0.5 mg/L indole-3-acetic acid (IAA).

Plantlets with well-developed roots were transferred to pots containing vermiculite and wetted with Hoagland’s nutrient solution (Hoagland & Arnon, 1938). Pots were covered with a beaker which was gradually removed until exposing the plants completely to ambient conditions. Self-fertilization and formation of grains were attained in the greenhouse. Ploidy level was verified in root tip squashes. Haploid plants had their crowns dipped into a solution of 0.25% colchicine and 2% dimethyl sulfoxide (DMSO) during 4 hours for doubling of the chromosomes.

The following traits were evaluated: AR/AI= (number of responsive anthers/number of plated anthers) x 100; PV/TP= (number of green plants/ total number of plants) x 100; PV/AR=(number of green plants/number of responsive anthers) x 100; TP/AR= (total number of plants/ number of responsive anthers) x 100; PV/AI= (number of green plants/ number of plated anthers) x 100; TP/AI= (total number of plants/ number of plated anthers) x 100.

Analysis of variance was performed on transformed data. Variable AR/AI was transformed using the equation y’=y0.22

while variable PV/AI was transformed with log(y+0.005). The others were transformed with log(y+0.02). Media were compared by Duncan’s multiple range test, at 5% probability level. Statistical analysis were done using proc GLM (SAS Institute, 1991) and ONEWAY (Nie et al., 1975). Association between genotypes and ploidy level was evaluated by correspondence analysis.

RESULTS AND DISCUSSION

Variance analysis showed a significant effect of culture media and genotype on the frequency of responsive anthers (AR/AI) (Table 1). Nevertheless, effects of these two factors are not independent, which means that genotypes behave differently in every culture medium and both culture media produce different effects, depending on the genotype. Percentage of responsive anthers on MS medium ranged between 0.21 and 15.55%, with an average of 6.39%. On the other hand, an average of 30.32% of the anthers cultured on N6 medium were responsive, with a minimum of 2.70% and a maximum of 45.18%. The superiority of N6 medium was statistically significant for most genotypes (Table 2). In this medium the potentialities of every genotype were expressed, allowing the identification of genotypes with high and low androgenetic capacities. Conversely, most of the structures formed on MS medium had a callus-like appearance, making regeneration more difficult. It was not possible to clearly differentiate genotypes due to a general deleterious effect.

TABLE 1. Analysis of variance for frequency of responsive anthers in barley anther culture.

Source of AR/AI1

Variation D.F. M.S.

Genotype 10 1.1879***Culture medium 1 10.8476***Genotype X medium 10 0.1908*Error 418 0.08311 AR/AI= (number of responsive anthers/ number of plated anthers) x

100; D.F.= degrees of freedom; M.S.= mean square; data were transformed using the equation y’=y0.22.

* = significant at p=0.05 and *** = significant at p= 0.001.

TABLE 2. Frequency (%) of responsive anthers on anther culture of eleven hybrid barley genotypes in two culture media.

Genotype Culture media1

N6 MS

AF 279 / PFC 9104 34.20 a A 4.18 bcd BBR 2 / MN 607 9.24 b A 0.39 de ABR 2 / PFC 9104 44.58 a A 9.82 abc BMN 607 / PFC 9104 29.75 a A 10.72 a ADEFRA / MN 607 2.70 b A 0.21 e AMN 656 / PFC 9104 29.22 a A 9.07 abc BMN 668 / PFC 9104 39.61 a A 15.55 a BPFC 85107 / PFC 9104 36.55 a A 2.40 cde BPFC 86104 / PFC 9104 32.33 a A 6.05 bc BPFC 9104 / PFC 9134 30.14 a A 4.03 bcd BPFC 9112 / PFC 88154 45.18 a A 7.82 ab B1 N6 (CHU, 1981); MS (MURASHIGE & SKOOG, 1962); VALUES

FOLLOWED BY THE SAME LETTER IN COLUMNS (MINUSCULE) OR ROWS (UPPER CASE) DO NOT DIFFER BY DUNCAN’S MULTIPLE RANGE TEST AT THE LEVEL OF 5%.

When compared with N6, MS medium has a greater amount of total nitrogen and ammonium (Grimes & Hodges, 1990). According to Halperin & Wetherell (1965), the proportion between reduced and oxidized inorganic nitrogen plays an important role in the induction and differentiation of plant cells in vitro. Many papers showed that androgenesis is hampered by high levels of ammonium. In rice the ratio NH4+/NO3-

affected callus induction, plantlet regeneration, the response to organic nitrogen, and the sensibility to 2,4-D (Grimes & Hodges, 1990). The influence got stronger as culture advanced, affecting mainly embryoid development and plantlet regeneration. Clapham (1973) observed that the frequency of responsive anthers was higher when the amount of NH4+ in culture medium was reduced to one tenth. According to Olsen (1987), high levels of ammonium induced the formation of callus that turned necrotic after a while in response to sub optimal culture conditions. Addition of glutamine and reduction of the amount of ammonium nitrate promoted the formation of embryoids that were able to regenerate green plantlets at higher frequencies. Modhorst &

Lörz (1993), stated that the poor development of microspores in media containing just NH4+ or with a high ratio NH4+/NO3- is caused by ammonium toxicity. It is also possible that the effects observed with the addition of ammonium to the culture medium may be caused by a change in cytokinin biosynthesis (Halperin & Wetherell, 1965; Weissman, 1972a, 1972b; Darral & Wareing, 1981; Beevers & Hageman, 1983; Mercier & Kerbauy, 1991).

While total frequencies of regeneration (TP/AR) was not significantly different between genotypes, analysis of variance showed that different genotypes had different frequencies of green plant formation (PV/AR) (Table 3). The greater capacity of some genotypes to originate green plantlets could be caused by differences in the ratio of green/total plantlets (PV/TP). Genotypes that regenerated many albino showed a low frequency of green plantlets (DEFRA/MN 607 and PFC 86104/PFC 9104) (Table 4). According to Logue et al. (1993), the ability to regenerate a great number of green plants depends on the reduction of albinism. Knudsen et al. (1989) believe that proportion of green plants and total frequency of regeneration are determined by different genetic traits. This explains why regeneration of green plants is more influenced by the ratio green/albino than by total frequency of regeneration. According to Knudsen et al. (1989) and Larsen et al. (1991), the proportion of green plants is strongly genotype dependent and has a great heritability. Logue et al. (1993) showed that the ratio green/al-bino remained constant even when donor plants were cultivated under totally different environments, like Great Britain and Australia.

TABLE 3. Analysis of variance for regeneration of plants in barley anther culture1.

Source of PV/TP PV/AR TP/AR PV/AI TP/AI

Variation D.F. M.S. D.F. M.S. D.F. M.S. D.F. M.S. D.F. M.S.

Genotype 10 0.2300* 10 0.5178** 10 0.3030ns 9 0.6331* 10 0.4727**Error 127 0.1179 166 0.2103 166 0.2738 199 0.3095 219 0.15921 PV/TP = (number of green plants/total number of plants) x 100; PV/AR = (number of green plants/number of responsive anthers) x 100;

TP/AR = (total number of plants/number of responsive anthers) x 100; PV/AI = (number of green plants/number of plated anthers) x 100;TP/AI = (total number of plants/number of plated anthers) x 100; D.F.= degrees of freedom, M.S.= mean square; variable PV/AI was transformed using the equation log(y+0.005); the others were transformed with log(y+0.02).

* = significant at p=0.05; ** = significant at 0.01; ns = not significant.

TABLE 4. Plants regenerated (observed and estima-ted values) (%) by a 100 responsive anthers from eleven hybrid barley geno-types cultured in N6 medium1.

Genotype PV/TP PV/AR TP/AR

Observed Observed Estimated Observed Estimated

AF 279 / PFC 9104 35.54 abc 5.44 (2.33) bc 12.94 (6.75) aBR 2 / MN 607 59.31 a 33.33 (13.13) a 55.65 (22.44) aBR 2 / PFC 9104 56.70 abc 14.03 (8.05) ab 21.94 (15.59) aMN 607 / PFC 9104 57.71 ab 14.03 (5.76) abc 25.12 (9.40) aDEFRA / MN 607 20.00 bc 4.29 (1.27) c 47.14 (14.42) aMN 656 / PFC 9104 45.92 abc 8.74 (4.05) abc 18.54 (9.24) aMN 668 / PFC 9104 45.36 abc 14.54 (5.37) abc 27.74 (12.85) aPFC 85107 / PFC 9104 39.07 abc 2.99 (1.87) bc 12.51 (5.76) aPFC 86104 / PFC 9104 19.37 c 2.90 (1.71) c 14.83 (10.41) aPFC 9104 / PFC 9134 60.80 a 12.60 (7.81) ab 21.41 (14.96) aPFC 9112 / PFC 88154 49.56 abc 5.07 (3.14) bc 11.18 (6.63) a1 PV/TP= (number of green plants/total number of plants) x 100;

PV/AR= (number of green plants/number of responsive anthers) x 100; TP/AR= (total number of plants/number of responsive anthers) x 100; statistical analysis were performed on estimated values (between paren thesis); values followed by the same letter in columns do not differ byDuncan’s multiple range test at the level of 5%.

In barley anther culture final productivity (PV/AI and TP/AI) is the result of a balance between induction of androgenetic structures and regeneration of green plants. Two groups of genotypes were identified (Table 5): one with a high embryogenic potential (PFC 9112/PFC 88154 and BR 2/PFC 9104) and the other with a great capacity to regenerate green plants (BR 2/MN 607). These results confirm the hypothesis that androgenetic capacity may result from two independent processes: induction and regeneration. These two

processes are controlled by different genetic mechanisms that may not be present together. The best genotypes are those combining a high frequency of induction with a reasonable frequency of regeneration. Populations derived from crosses with the line PFC 9104 (BR 2/PFC 9104, MN 668/PFC 9104, and PFC 9104/PFC 9134) showed high androgenetic capacity. This line descends from Igri, which extraordinary androgenetic capacity has been recognized and used in many studies (Foroughi-Wehr & Friedt, 1984; Olsen, 1987; Ziauddin et al., 1990). According to Larsen et al. (1991), it is common for hybrids to present levels of response equal or superior to their parents.

TABLE 5. Plants produced by a 100 plated anthers from eleven hybrid barley genotypes cultured in N6 medium1.

Genotype PV/AI TP/AI

Observed Estimated Observed Estimated

AF 279 / PFC 9104 3.04 (0.71) abc 5.42 (2.63) abBR 2 / MN 607 5.41 (0.46) c 7.58 (1.43) bcBR 2 / PFC 9104 4.40 (2.34) a 8.27 (5.57) aMN 607 / PFC 9104 4.21 (1.46) abc 8.43 (3.55) abDEFRA / MN 607 0.24 − 0.93 (0.45) cMN 656 / PFC 9104 2.91 (1.01) abc 5.69 (3.27) abMN 668 / PFC 9104 8.26 (1.71) abc 13.65 (5.71) aPFC 85107 / PFC 9104 0.97 (0.51) bc 4.04 (2.15) abcPFC 86104 / PFC 9104 1.20 (0.55) bc 5.20 (3.60) abPFC 9104 / PFC 9134 4.34 (2.03) ab 7.24 (4.20) abPFC 9112 / PFC 88154 2.45 (1.26) abc 5.03 (3.37) ab1 PV/A.I: (number of green plants/number of plated anthers) x 100;

TP/AI: (total number of plants/number of plated anthers) x 100; statistical analysis were performed on estimated values (between parenthesis); values followed by the same letter in columns do not differ by Duncan’s multiple range test at the level of 5%.

Barley anther culture usually shows a high percentage of spontaneous doubling of the chromosomes making unnecessary to treat plants with colchicine in order to obtain doubled haploids (Foroughi-Wehr & Friedt, 1984; Olsen, 1987; Luckett & Smithard, 1992). A high frequency of diploids was observed (Table 6), but an association between ploidy level and genotype was detected by correspondence analysis (Fig. 1). Finnie et al. (1989; 1991), Bjørnstad et al. (1993) and Logue et al. (1993) found similar differences in the frequency of spontaneous doubling using cultivars or F1 populations. Evidence suggests that there may be some level of genetic control over the ploidy of plants obtained by anther culture. In this case, there was a high frequency of polyploids associated with cultivar MN 607 (MN 607/PFC 9104 and DEFRA/MN 607) suggesting that the presence of this cultivar favors polyploidization. On the other hand, BR 2/MN 607, BR 2/PFC 9104 and PFC 9112/PFC 88154 presented a low frequency of spontaneous doubling. At the same time, these crosses had a high capacity to regenerate green plants. This may lead to the conclusion that polyploidization has a negative effect over regeneration of plants.

TABLE 6. Ploidy level of plants regenerated by barley anther culture.

Genotype Ploidy level

Haploid Diploid Poliploid

AF 279 / PFC 9104 3 6 0BR 2 / MN 607 19 7 0BR 2 / PFC 9104 16 9 0MN 607 / PFC 9104 3 2 3DEFRA / MN 607 0 0 1MN 656 / PFC 9104 10 13 0MN 668 / PFC 9104 14 55 4PFC 85107 / PFC 9104 0 2 0PFC 86104 / PFC 9104 1 3 0PFC 9104 / PFC 9134 4 5 0PFC 9112 / PFC 88154 8 4 0

Total 78 106 8

FIG. 1. Correspondence analysis between genotype and ploidy level (103: AF 279 / PFC 9104; 105: BR 2 / MN 607; 108: BR 2 / PFC 9104; 109: MN 607 / PFC 9104; 111: DEFRA / MN 607; 117: MN 656 / PFC 9104; 123: MN 668 /PFC 9104; 125:PFC 85107/PFC 9104; 126: PFC 86104/PFC 9104; 135: PFC 9104/PFC 9134; 138: PFC 9112/PFC 88154).

According to Luckett & Smithard (1992), a productivity of 10 plants/100 plated anthers allows the production of 100 doubled haploid lines from 50 different crosses by a single worker during one year. In Europe anther culture has been integrated to barley breeding programs (Kühlmann & Foroughi--Wehr, 1989). Using different F1 populations an average of 3.52 and a maximum of 8.26 green plants/100 plated anthers was obtained. These results are similar or superior to those obtained by Foroughi--Wehr & Friedt (1984), Devaux (1987), Olsen (1987), Powell et al. (1988), Finnie et al. (1989), Knudsen et al. (1989), Kühlmann & Foroughi-Wehr (1989), Luckett & Smithard (1992), and Kintzios & Fischbeck (1994). Obtaining a reasonable number of doubled haploids bearing genes that confer productivity and adaptation to local environmental conditions makes anther culture an useful tool to be used in Brazilian barley breeding programs.

CONCLUSIONS

1. Presence of line PFC 9104 confers a high androgenetic capacity.2. Medium N6 is superior to MS for barley anther culture allowing the obtainment of doubled haploids

from different lines.

ACKNOWLEDGMENTS

To the direction of Embrapa-Centro Nacional de Pesquisa de Trigo (CNPT), for providing material and human resources; to Gelsi Galon, for his technical assistance; to Dr. João Riboldi and Álvaro Vigo from Universidade Federal do Rio Grande do Sul, for the statistical analysis; to FINEP and FAPERGS, for financial support. Special thanks from Ana Lucia Stival da Silva for Master’s fellowship from CNPq.

REFERENCES

BEEVERS, L.; HAGEMAN, R.H. Uptake and reduction of nitrate in bacteria and higher plants. In: LAUCHLI, A.; BIELESKY, R.L. (Eds.). Inorganic plant nutrition. Berlin: Springer-Verlag, 1983. p.351-375.

BJØRNSTAD, Å.; SKINNES, H.; THORESEN, K. Comparison between doubled haploid lines produced by anther culture, the Hordeum bulbosum-method and lines produced by single seed descent in barley crosses. Euphytica, Dordrecht, v.66, p.135-144, 1993.

105 138 108135 117 103 126

123

125

109

111

Haploid

Diploid

Poliploid

CAI, Q.; SZAREJKO, I.; POLOK, K.; MALUSZYNSKI, M. The effect of sugars and growth regulators on embryoid formation and plant regeneration from barley anther culture. Plant Breeding, Berlin, v.109, p.218-226, 1992.

CHU, C.-C. The N6 medium and its applications to anther culture of cereal crops. In: SYMPOSIUM ON PLANT TISSUE CULTURE, 1978, Peking. Proceedings... Boston: Pitman Advanced Pub. Program, 1981. p.43-50.

CLAPHAM, D. Haploid Hordeum plants from anthers in vitro. Zeitschrift für Pflanzenzuechtung, v.69, p.142-155, 1973.

DARRAL, N.M.; WAREING, P.F. The effect of nitrogen nutrition on cytokinin activity and free amino acids in Betula pendula Roth. and Acer pseudoplatanus L. Journal of Experimental Botany, Oxford, v.32, p.369-379, 1981.

DEVAUX, P. Comparison of anther culture and Hordeum bulbosum method for the production of double haploids in winter barley. I. Production of green plants. Plant Breeding, Berlin, v.98, p.215-219, 1987.

FERNANDES, M.I.B. de M. Perspectivas da biotecnologia para o melhoramento de plantas. Pesquisa Agropecuária Brasileira, Brasília, v.22, n.9/10, p.881-896, set./out. 1987.

FERNANDES, M.I.B. de M. Obtenção de plantas haplóides através da cultura de anteras. In: TORRES, A.C.; CALDAS, L.S. (Eds.). Técnicas e aplicações da cultura de tecidos de plantas. Brasília: ABCTP/Embrapa-CNPH, 1990. p.311-332.

FERNANDES, M.I.B. de M.; CAETANO, V.R.; PICARD, E. Biotecnologia no melhoramento do trigo (Triticum aestivum): produção de plantas haplóides através da cultura de anteras “in vitro” para acelerar o processo de criação de cultivares, no Centro Nacional de Pesquisa de Trigo/Embrapa. Ciência e Cultura, São Paulo, v.42, p.472-4732, l990.

FINNIE, S.J.; FORSTER, B.P.; CHALMERS, K.J.; DYER, A.F.; WAUGH, R.; POWELL, W. Genetic stability of microspore-derived doubled haploids of barley: a cytological, biochemical and molecular study. Genome, Ottawa, v.34, p.923-928, 1991.

FINNIE, S.J.; POWELL, W.; DYER, A.F. The effect of carbohydrate composition and concentration on anther culture response in barley (Hordeum vulgare L.). Plant Breeding, Berlin, v.103, p.110-118, 1989.

FOROUGHI-WEHR, B.; FRIEDT, W. Rapid production of recombinant barley yellow mosaic virus resistant Hordeum vulgare lines by anther culture. Theoretical and Applied Genetics, Berlin, v.67, p.377-382, 1984.

GRIMES, H.D.; HODGES, T.K. The inorganic NO3-:NH4+ ratio influences plant regeneration and auxin sensitivity in primary callus derived from immature embryos of indica rice (Oryza sativa L.). Journal of Plant Physiology, Stuttgart, v.136, p.362-367, 1990.

GUHA, S.; MAHESHWARI, S.C. Cell division and differentiation of embryos in the pollen grains of Datura in vitro. Nature, New York, v.212, p.97-98, 1966.

HALPERIN, W.; WETHERELL, D.F. Ammonium requirement for embryogenesis “in vitro”. Nature, New York, v.205, p.519-520, 1965.

HOAGLAND, D.R.; ARNON, D.I. The water-culture method for growing plants without soil. Berkeley: Univ. of California, 1938. 39p.

HUANG, B.; SUNDERLAND, N. Temperature-stress pretreatment in barley anther culture. Annals of Botany, London, v.49, p.77-88, 1982.

KINTZIOS, S.; FISCHBECK, G. Anther culture response of Hordeum spontaneum - derived winter barley lines. Plant Cell, Tissue and Organ Culture, The Hague, v.37, p.165-170, 1994.

KNUDSEN, S.; DUE, I.K.; ANDERSEN, S.B. Components of response in barley anther culture. Plant Breeding, Berlin, v.103, p.241-246, 1989.

KÜHLMANN, U.; FOROUGHI-WEHR, B. Production of doubled haploid lines in frequencies sufficient for barley breeding programs. Plant Cell Reports, Berlin, v.8, p.78-81, 1989.

LARSEN, E.T.; TUVESSON, I.K.D.; ANDERSEN, S.B. Nuclear genes affecting percentage of green plants in barley (Hordeum vulgare L.) anther culture. Theoretical and Applied Genetics, Berlin, v.82, p.417-420, 1991.

LOGUE, S.J.; GILES, L.C.; SPARROW, D.H.B. Genotype and environment strongly influence barley anther culture response using Australian genotypes. Australian Journal of Botany, Melbourne, v.41, p.227-236, 1993.

LUCKETT, D.J.; SMITHARD, R.A. Doubled haploid production by anther culture for Australian barley breeding. Australian Journal of Agricultural Research, Melbourne, v.43, p.67-68, 1992.

MERCIER, H.; KERBAUY, G.B. Effects of nitrogen source on growth rates and levels of endogenous cytokinins and chlorophyll in protocorms of Epidendrum fulgens. Journal of Plant Physiology, Stuttgart, v.138, p.195-199, 1991.

MODHORST, A.P.; LÖRZ, H. Embryogenesis and development of isolated barley (Hordeum vulgare L.) microspores are influenced by the amount and composition of nitrogen sources in culture media. Plant Physiology, Rockville, v.142, p.485-492, 1993.

MURASHIGE, T.; SKOOG, F. A revised medium for rapid growth and bio assays with tobacco tissue cultures. Physiologia Plantarum, Copenhagen, v.15, p.473-497, 1962.

NIE, N.H.; HULL, C.H.; JENKINS, J.G.; STEINBRENNER, K.; BENT, D.H. SPSS: Statistical Package for the Social Sciences. 2.ed. New York: McGraw-Hill, 1975. 675p.

OLSEN, F.L. Induction of microspore embryogenesis in cultured anthers of Hordeum vulgare: the effects of ammonium nitrate, glutamine and asparagine as nitrogen sources. Carlsberg Research Communications, Copenhagen, v.52, p.393-404, 1987.

POWELL, W.; BORRINO, E.M.; GOODALL, V. The effect of anther orientation on microspore-derived plant production in barley (Hordeum vulgare L.). Euphytica, Dordrecht, v.38, p.159-163, 1988.

SAS Institute Inc. SAS/GRAPH Software: usage. Version 6. Cary, NC, 1991. 887p.SHELL INTERNATIONALE RESEARCH MAATSCHAPPIJ B.V.; HUNTER, C.P. Plant generation method. Int. Cl.

4: AO1G 7/00 GB 0 245 898 A2 19 nov. 1987. European Patent Office. Bulletin 87/47. 8p.WEISSMAN, G.S. Influence of ammonium and nitrate nutrition on enzymatic activity in soybean and sunflower. Plant

Physiology, Rockville, v.49, p.138-141, 1972a.

WEISSMAN, G.S. Influence of ammonium and nitrate nutrition on the pyridine and adenine nucleotides of soybean and sunflower. Plant Physiology, Rockville, v.49, p.142-145, 1972b.

WENZEL, G. Application of unconventional techniques in classical plant production. In: FOWLER, M.W.; WARREN, G.S.; MOO-YOUNG, M. (Eds.). Plant biotechnology: comprehensive biotechnology. Oxford: Pergamon Press, 1992. v.2, p.259-281.

ZIAUDDIN, A.; SIMION, E.; KASHA, K.J. Improved plant regeneration from shed microspore culture in barley (Hordeum vulgare L.) cv. Igri. Plant Cell Reports, Berlin, v.9, p.69-72, 1990.